On the Modelling and Forecasting of Multivariate Realized Volatility: Generalized Heterogeneous Autoregressive (GHAR) Model

Recent multivariate extensions of the popular heterogeneous autoregressive model (HAR) for realized volatility leave substantial information unmodelled in residuals. We propose to employ a system of seemingly unrelated regressions to model and forecast a realized covariance matrix to capture this information. We find that the newly proposed generalized heterogeneous autoregressive (GHAR) model outperforms competing approaches in terms of economic gains, providing better mean–variance trade‐off, while, in terms of statistical precision, GHAR is not substantially dominated by any other model. Our results provide a comprehensive comparison of the performance when realized covariance, subsampled realized covariance and multivariate realized kernel estimators are used. We study the contribution of the estimators across different sampling frequencies, and show that the multivariate realized kernel and subsampled realized covariance estimators deliver further gains compared to realized covariance estimated on a 5‐minute frequency. In order to show economic and statistical gains, a portfolio of various sizes is used. Copyright © 2016 John Wiley & Sons, Ltd.     

Cross‐Section Stock Return and Implied Covariance between Jump and Diffusive Volatility

I examine the information content of option‐implied covariance between jumps and diffusive risk in the cross‐sectional variation in future returns. This paper documents that the difference between realized volatility and implied covariance (RV‐ICov) can predict future returns. The results show a significant and negative association of expected return and realized volatility–implied covariance spread in both the portfolio level analysis and cross‐sectional regression study. A trading strategy of buying a portfolio with the lowest RV‐ICov quintile portfolio and selling with the highest one generates positive and significant returns. This RV‐Cov anomaly is robust to controlling for size, book‐to‐market value, liquidity and systematic risk proportion. Copyright © 2015 John Wiley & Sons, Ltd.     

对称熵损失下两个指数总体均值的序约束估计

在对称熵损失下, 讨论了样本容量相等时, 两个指数总体均值λ_i(i=1,2)的约束极大似然估计i的险, 其中约束为λ₁≤λ₂. 证明了λ₁与λ₂具有比经典极大似然估计X₁与X₂ 更小的风险, 并给出了当λ₂/λ₁→∞和n→∞时,λ_i对X_i(i=1,2)渐近功效e(λ_i,X_i)的值.     

Linex损失下Rayleigh分布参数倒数的Bayes估计

在Linex损失函数L(θ,δ)=ec(δ-θ)-c(δ-θ)-1,c>0下,给出Rayleigh分布的尺度参数倒数的唯一Bayes估计δB(X)=-1/clnE(e-cθ│X)=(n α)/cln(1 c/(λ T)),多层Byaes估计δ∧B(X)=-1/cln,和容许性估计的一般形式Sln(1 c/(d T)).     

几个导出模型的BLU估计间的关系的研究

首先给出分块奇异模型的一种新的导出模型,得出其M2X1β的BLU估计;讨论分块模型与几个导出模型的BLU估计之间的关系,给出这几个分块导出模型的M2X1β的BLU估计相等的充要条件.     

半相依回归系统参数的c-k型改进估计

针对设计矩阵Xi呈病态时的半相依线性回归系统,提出了系统参数iβ的一种c-k型改进估计,并证明了这种估计在均方误差意义下的若干优良性质.     

广义系统多传感器分布式融合降阶Kalman滤波器

对于带多传感器的广义线性离散随机系统,应用奇异值分解,将其变换为等价的两个降阶多传感器子系统,提出了基于变换后的状态融合器构造原始状态融合器的新的融合方法。应用Kalman滤波方法,在线性最小方差按矩阵加权、按对角阵加权和按标量加权融合准则下,分别提出了三种最优加权融合降阶广义Kalman滤波器。可统一处理融合滤波、平滑和预报问题。可减少计算负担和改善局部滤波精度。证明了三种融合器和局部估值器之间的精度关系。为了计算最优加权。提出了局部滤波误差协方差阵的计算公式。一个Monte Carlo仿真例子说明了其有效性。     

辨识动态系统噪声方差Q和R的新方法

对于带未知噪声方差的线性离散定常随机系统,引入左素分解可得到一个新的观测过程,它用两个滑动平均（MA）过程之和表示。用解相关函数矩阵方程组得到了噪声方差Q和R的估值器,进而基于新的观测过程的采样相关函数及其遍历性可得到噪声方差Q和R的强一致估计。算法简单,便于实时应用。一个目标跟踪系统的仿真例子说明了其有效性。     

基于KULLBACK信息融合方法的系统可靠性评估

在产品具有多源验前信息的情形下,讨论冷贮备系统的可靠性评估问题,运用Kullback信息作为分布之间距离的度量,在Kullback信息的融合准则下对多个先验分布进行融合。并以融合后的先验分布作为产品的最终验前分布,对冷贮备系统的可靠性指标进行Bayes估计。最后进行的计算机随机模拟结果表明,文中所提出的方法合理且便于应用。     

台湾休闲农业的成功经验及对大陆的启示

休闲农业在台湾发展较早,现已具有相当的规模,并取得了良好的社会、经济和生态效益,是台湾农业转型的新走向。介绍了台湾休闲农业的发展概况、取得的成功经验及其对大陆农业发展带来的一些启示。